Dear Lammps Users,
I am simulating Al2O3 films which are periodic in x and y direction and non-
periodic in z direction. I have placed vacuum in z direction above and below
the crystal.
I perform simulations for different thickness of films. I well equilibrate
the crystal using NPT ensemble. I determine the total energy.
I observe that total energy of one or two thickness of films to be
exceptionally high. I am varying thickness in z direction and I get
following results after equilibration:
Thickness Total Energy after equilibration
2 nm -6.3196651 ev/atom
4 nm -6.3196691 ev/atom
6 nm -6.2209879 ev/atom
8nm -6.2873554 ev/atom
10 nm -6.3122231 ev/atom
I observe for 6 nm thick film energy to be maximum. What could be the
possible cause?
impossible to say. can be lots of things: inadequate simulation
protocol, unsuitable potential parameters, bad/inconsistent choice of
initial geometry, genuine structural instability, insufficient
equilibration (i.e. some of your systems are stuck in some meta stable
state, while others can relax into a different, lower energy
structure), incorrect analysis, etc.
there also is very little additional information compared to your
previous post. more importantly, this is turning into a discussion
topic that is in no way specific to using LAMMPS, but rather about how
to perform and validate simulations in general. that is something, you
need to discuss with your adviser and/or more experienced
colleagues/collaborators. please understand that the purpose of this
mailing list is to help people with LAMMPS specific problems, not to
advise them on how to do their research and explain their results to
them.
I read on the forum that high energies can be a result of periodic images of
atoms getting very close. Does it happen for case of 6 nm?
you yourself are in the best position to determine that, as you have
access to your trajectory data and can easily verify whether this
condition applies. if that is indeed the case, however, it would cast
a large doubt on your overall simulation protocol, as with a correctly
performed equilibration/relaxation there should be no close contacts
via PBC.
axel.